Emotions play a critical role in organizing human experience and behavior, and emotion dysregulation lies at the heart of psychopathology and functional impairment across disorders. To measure and understand emotion dysregulation, advances in understanding the fundamentals of how the brain generates and represents emotional states are vitally needed. This proposal develops and validates models of the brain representations that give rise to emotional states in naturalistic, narrative contexts. This will provide normative models of emotion to ground future translational studies, measurement models for specific emotional brain representations, and targets for interventions. We combine Functional Magnetic Resonance Imaging (fMRI), multi-dimensional measures of behavior, and pattern recognition techniques to develop models of brain activity that characterize and differentiate discrete categories of emotion experience (joy, anger, sadness, pride, and others) and blends of emotion. We place particular emphasis on the predictive validity (sensitivity and specificity) and generalizability of these models across sensory modalities, evaluative judgments, contextual narratives, and populations. We elicit emotional experiences in an ecologically valid paradigm using narratives (stories) experienced via listening, reading, or watching video. We measure multiple types of emotional experience in parallel with fMRI, using innovative collaborative filtering approaches to infer continuous moment-by-moment experience. The resulting brain models of specific emotion categories afford several potentially transformative advantages. Such models can (a) provide insight into which systems are necessary and sufficient for emotion generation (Aim 1); (b) be shared and tested across studies, allowing us to evaluate their generalizability across contexts (Aim 2); and (c) provide targets for psychological and neurological interventions (Aim 3). Six experiments focus on developing and validating emotional brain representations that are generalizable across individuals, research sites (Dartmouth and Colorado), and populations (college students and more diverse community samples). Expt. 1 develops models that predict the intensity of discrete emotional states. Expts. 2-4 establish the context sensitivity and generalizability of these. Expt. 2 examines the role of evaluative judgments in shaping emotional experience. Expt. 3 assesses the impact of background contextual narratives. Expt. 4 evaluates the role of sensory processing in emotion representations. Expts. 5-6 establish whether or not the brain models mediate emotional experiences. Expt. 5 uses cognitive appraisal and Expt. 6 uses real-time fMRI neurofeedback to manipulate emotion category-specific brain representations, testing for causal effects of these psychological and brain manipulations on emotional experience. Together, these studies will yield generalizable models of the dynamic brain patterns underlying specific emotional experiences. Such models could transform clinical research by allowing investigators to test emotion- focused interventions and assess emotion-related risk factors, permitting early detection and intervention.